JMIR human factors · 2026
Large Language Model-Based Simplification of Digital Therapeutics Explanations for Insomnia and Nicotine Dependence: Two Randomized Online Experiments.
СтатьяРандомизированное исследованиеПоддержка исследования: вне правительства СШАНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Seo JunYoung, Yook Moses, Kim Dai Jin, Lee JunHee, Yoo Jae Hyun, Byeon GiHwan, Choi In Young
Аннотация
BACKGROUND: Digital therapeutics (DTx) are evidence-based software interventions with the potential to treat health conditions. However, uptake remains limited by low public awareness and overly complex patient education materials that exceed recommended readability levels. Large language models (LLMs) may simplify such content; however, their effects on users' understanding have not been empirically demonstrated.
OBJECTIVE: This study aimed to examine whether LLM-based simplification of DTx explanatory materials enhances perceived understanding and subjective evaluations of readability, clarity, and comprehensibility compared with manufacturer-provided documents.
METHODS: We developed a simplification tool using the GPT-4o application programming interface (API), configured for deterministic outputs and guided by structured readability instructions. Original DTx explanatory materials about insomnia and nicotine dependence were obtained from manufacturers and transformed into simplified versions. Two randomized, between-subject online experiments were conducted (n=1000, with 500 participants in each experiment). Participants were stratified by age and sex and screened for relevance (Insomnia Severity Index ≥8 for the insomnia experiment and smoking ≥5 cigarettes per day for the nicotine dependence experiment). Within each experiment, participants were randomly assigned to review either the original or the LLM-simplified explanatory material. Perceived understanding was assessed before and after exposure. Postexposure evaluations of ease, clarity, and comprehensibility were also collected.
RESULTS: Repeated measures ANOVA revealed significant group×time interaction effects on perceived understanding in both experiments: insomnia (F1,498=24.8; P<.001) and nicotine dependence (F1,498=14.1; P<.001), with greater improvements observed in the LLM-simplified groups. Mann-Whitney U tests further showed that LLM-simplified explanations were rated as easier, clearer, and more comprehensible than the original versions in both experiments (all P<.05), with small to moderate effect sizes (r=0.11-0.24).
CONCLUSIONS: Compared with manufacturer-provided original materials, LLM-simplified DTx explanations led to greater improvements in perceived understanding and subjective evaluations of readability among lay audiences, even after a single exposure. This finding highlights the potential scalability of LLM-based simplification as a strategy to improve the perceived accessibility of health information for lay audiences. Integrating such tools into patient education may enhance how lay audiences perceive and engage with DTx, although further research using objective comprehension measures is needed to confirm these benefits.
OBJECTIVE: This study aimed to examine whether LLM-based simplification of DTx explanatory materials enhances perceived understanding and subjective evaluations of readability, clarity, and comprehensibility compared with manufacturer-provided documents.
METHODS: We developed a simplification tool using the GPT-4o application programming interface (API), configured for deterministic outputs and guided by structured readability instructions. Original DTx explanatory materials about insomnia and nicotine dependence were obtained from manufacturers and transformed into simplified versions. Two randomized, between-subject online experiments were conducted (n=1000, with 500 participants in each experiment). Participants were stratified by age and sex and screened for relevance (Insomnia Severity Index ≥8 for the insomnia experiment and smoking ≥5 cigarettes per day for the nicotine dependence experiment). Within each experiment, participants were randomly assigned to review either the original or the LLM-simplified explanatory material. Perceived understanding was assessed before and after exposure. Postexposure evaluations of ease, clarity, and comprehensibility were also collected.
RESULTS: Repeated measures ANOVA revealed significant group×time interaction effects on perceived understanding in both experiments: insomnia (F1,498=24.8; P<.001) and nicotine dependence (F1,498=14.1; P<.001), with greater improvements observed in the LLM-simplified groups. Mann-Whitney U tests further showed that LLM-simplified explanations were rated as easier, clearer, and more comprehensible than the original versions in both experiments (all P<.05), with small to moderate effect sizes (r=0.11-0.24).
CONCLUSIONS: Compared with manufacturer-provided original materials, LLM-simplified DTx explanations led to greater improvements in perceived understanding and subjective evaluations of readability among lay audiences, even after a single exposure. This finding highlights the potential scalability of LLM-based simplification as a strategy to improve the perceived accessibility of health information for lay audiences. Integrating such tools into patient education may enhance how lay audiences perceive and engage with DTx, although further research using objective comprehension measures is needed to confirm these benefits.